From mboxrd@z Thu Jan 1 00:00:00 1970 From: tupeng212 Subject: Re: [PATCH, RFC v2] x86/HVM: assorted RTC emulation adjustments (was Re: Big Bug:Time in VM goes slower...) Date: Thu, 16 Aug 2012 21:43:32 +0800 Message-ID: <201208162143289686654@gmail.com> References: <502A3BBC0200007800094B68@nat28.tlf.novell.com>, <2012081522045495397713@gmail.com> <2012081522121039050717@gmail.com>, <502CC9D702000078000955B9@nat28.tlf.novell.com> Reply-To: tupeng212 Mime-Version: 1.0 Content-Type: multipart/mixed; boundary="===============4223491905756042710==" Return-path: List-Unsubscribe: , List-Post: List-Help: List-Subscribe: , Sender: xen-devel-bounces@lists.xen.org Errors-To: xen-devel-bounces@lists.xen.org To: Jan Beulich Cc: Yang Z Zhang , Tim Deegan , Keir Fraser , xen-devel List-Id: xen-devel@lists.xenproject.org This is a multi-part message in MIME format. --===============4223491905756042710== Content-Type: multipart/alternative; boundary="----=_001_NextPart451220271581_=----" This is a multi-part message in MIME format. ------=_001_NextPart451220271581_=---- Content-Type: text/plain; charset="gb2312" Content-Transfer-Encoding: base64 RGVhciBKYW46DQpJIGRpZG4ndCBjaGVjayB0aGUgZmxpcHBpbmcgaW4gSlZNLCBidXQgZnJvbSB0 aGUgcHJpbnRpbmcgcGVyaW9kIGluIHhlbiwgdGhpcyBzZXR0aW5nIGhhcHBlbmVkKHVuaXQ6IG5z KToNCjk3NjU2MiwgOTc2NTYyLCA5NzY1NjIsIDk3NjU2MiwgMTU2MjUwMDAsIDk3NjU2MiwgOTc2 NTYyLCA5NzY1NjIsIDk3NjU2MiwgMTU2MjUwMDAgLi4uLi4NCmFuZCBzb21lb25lIHRvbGQgbWUg d2luZG93cyB1c2UgdGhlc2UgdHdvIHJhdGUvcGVyaW9kIG9ubHkuDQoNCmJlc2lkZXMsIEkgY2hl Y2tlZCBteSBmb3JtZXIgc2ltcGxlIHRlc3RlciB0aGlzIG1vcm5pbmcgYWZ0ZXIgaXQgcmFuIGZv ciBhIHdob2xlIG5pZ2h0LCBpdCBsYWdnZWQgbXVjaC4NCnNvIHRoZXJlIGV4aXN0cyBkaWZmZXJl bmNlIGJldHdlZW4gdmlydHVhbGl6YXRpb24gYW5kIHJlYWxpdHkuDQoNCg0KDQp0dXBlbmcyMTIN Cg0KRnJvbTogSmFuIEJldWxpY2gNCkRhdGU6IDIwMTItMDgtMTYgMTY6MjINClRvOiB0dXBlbmcy MTINCkNDOiBZYW5nIFogWmhhbmc7IHhlbi1kZXZlbDsgS2VpciBGcmFzZXI7IFRpbSBEZWVnYW4N ClN1YmplY3Q6IFJlOiBbWGVuLWRldmVsXSBbUEFUQ0gsIFJGQyB2Ml0geDg2L0hWTTogYXNzb3J0 ZWQgUlRDIGVtdWxhdGlvbiBhZGp1c3RtZW50cyAod2FzIFJlOiBCaWcgQnVnOlRpbWUgaW4gVk0g Z29lcyBzbG93ZXIuLi4pDQo+Pj4gT24gMTUuMDguMTIgYXQgMTY6MTIsIHR1cGVuZzIxMiA8dHVw ZW5nMjEyQGdtYWlsLmNvbT4gd3JvdGU6DQo+IFRoZSByZXN1bHRzIGZvciBtZSB3ZXJlIHRoZXNl Og0KPiAxIEluIG15IHJlYWwgYXBwbGljYXRpb24gZW52aXJvbm1lbnQsIGl0IHdvcmtlZCB2ZXJ5 IHdlbGwgaW4gdGhlIGZvcm1lciANCj4gNW1pbnMsIG11Y2ggYmV0dGVyIHRoYW4gYmVmb3JlLA0K PiAgYnV0IGF0IGxhc3QgaXQgbGFnZ2VkIGFnYWluLiBJIGRvbid0IGtub3cgd2hldGhlciBpdCBi ZWxvbmdzIHRvIHRoZSB0d28gDQo+IG1pc3NlZCBmdW5jdGlvbnMuIEkgbGFjayB0aGUgDQo+ICBh YmlsaXR5IHRvIGZpZ3VyZSB0aGVtIG91dC4NCg0KRGlkIHlvdSBjaGVjayB3aGV0aGVyIHBvc3Np Ymx5IHRoZSBndWVzdCBrZXJuZWwgc3RhcnRlZCBmbGlwcGluZw0KYmV0d2VlbiB0byByYXRlIHZh bHVlcz8gSWYgc28sIHRoYXQgaXMgc29tZXRoaW5nIHRoYXQgd2UgY291bGQNCmRlYWwgd2l0aCB0 b28gKHlldCBpdCBtaWdodCBiZSBhIGJpdCBpbnZvbHZlZCwgc28gSSdkIGxpa2UgdG8gYXZvaWQN CmdvaW5nIHRoYXQgcm91dGUgaWYgaXQgd291bGQgbGVhZCBub3doZXJlKS4NCg0KSmFu ------=_001_NextPart451220271581_=---- Content-Type: text/html; charset="gb2312" Content-Transfer-Encoding: quoted-printable

Dear Jan:

I didn't check the flipping in JVM, but from the printing period in x= en,=20 this setting happened(unit: ns):

976562, 976562, 976562, 976562, 15625000, 976562, 976562, 976562, 976= 562,=20 15625000 .....

and someone told me windows use these two rate/period only.

besides, I checked my former simple tester this morning after it ran = for a=20 whole night, it lagged much.

so there exists difference between virtualization and=20 reality.

tupeng212

From: Jan Beulich

Date: 2012-08-16 16:22

To: tupeng212<= /DIV>

CC: Yang Z Zhan= g;=20 xen-devel; Keir Fraser; Tim=20 Deegan

Subject: Re: [Xen-devel] [PATCH, RFC v2] x86/HVM: assorte= d RTC=20 emulation adjustments (was Re: Big Bug:Time in VM goes=20 slower...)

>>> On 15.08.12 at 16:12, tupeng212= <tupeng212@gmail.com> wrote:

> The results for me were these:

> 1 In my real application environme= nt, it worked very well in the former&n= bsp;

> 5mins, much better than before,

> but at last it lagged again.= I don't know whether it belongs to&nbs= p;the two

> missed functions. I lack the

> ability to figure them out.

Did you check whether possibly the gues= t kernel started flipping

between to rate values? If so, that&nbs= p;is something that we could

deal with too (yet it might be a&n= bsp;bit involved, so I'd like to avoid

going that route if it would lead = nowhere).

Jan

------=_001_NextPart451220271581_=------ --===============4223491905756042710== Content-Type: text/plain; charset="us-ascii" MIME-Version: 1.0 Content-Transfer-Encoding: 7bit Content-Disposition: inline _______________________________________________ Xen-devel mailing list Xen-devel@lists.xen.org http://lists.xen.org/xen-devel --===============4223491905756042710==--